Research Themes

Health＋Causal Inference


Analyzing Human Data to Build a Society
Where People Can Live Long and Healthy Lives!

EpidemiologyCausal InferenceGenomic InformationCancer RiskHealthy Life Expectancy

Using Data to Extend Healthy Life Expectancy

My area of expertise is epidemiology. Epidemiology is a field of study that identifies patterns in health-related events that occur in human populations—such as their types, frequency, and causes—and links those findings to practical countermeasures.
In particular, I focus on diet and health, and I have been involved in research such as: “Can people live longer by eating fermented soybean foods (natto, miso, etc.)?” and “Does diet affect sleep quality?”
To obtain research findings with a high level of evidence, it is necessary to follow a population over a long period of time, and the larger the population size, the more reliable the results become. Sometimes we collect data from a population of 100,000 people in Japan or 300,000 people across Asia, and data-science methods such as machine learning and causal inference＊ are indispensable for analyzing and interpreting those data.

Epidemiology: A public health field that collects and analyzes data on human health and connects the results to countermeasures.

Tips

Causal Inference
A statistical method for estimating the causal relationship between two things. Even if there are data showing that “people who ate natto lived longer,” verifying whether there is truly a causal relationship requires complex steps.

Epidemiological Research Using Image Analysis Technology

Traditionally, we used paper-based data such as questionnaires, formed hypotheses about the relationship between outcomes and contributing factors, and then conducted assessments (evaluation and appraisal). However, as image analysis technology advances, it may become possible to collect data by capturing images of participants’ meals even when they are in remote locations, and it may even become possible to infer participants’ emotions from those images.
In recent years, machine-learning methods have also been incorporated into statistical analyses, and computers are becoming able to predict the factors behind certain outcomes. When forming hypotheses about “outcomes and factors,” researchers used to anticipate possible factors and analyze them one by one. With advances in machine learning, however, it is becoming possible to have AI consider even “what could potentially be a factor.”
By closely integrating information engineering and data science, research on diet and health is expected to gain even greater depth and accuracy than before.

Data Processing Technology Applicable to Any Field

The process of collecting and processing massive amounts of data and then extracting insights from them is not limited to epidemiology; it is a basic process common to data science in general. It is a technology that can be applied across society, including industry, finance, the environment, and entertainment.
You may think that machine learning and data science sound difficult, but in fact they are technologies used surprisingly close to our daily lives—such as health and eating habits. Personal health data will be aggregated on mobile devices such as smartwatches, and scanning a supermarket product barcode will display a dinner recipe tailored to your physical condition that day—such things will become commonplace.
To reach that day, we need your energy, curiosity, and desire to improve. Let’s open up the future together at the Faculty of Informatics.

Analyzing the causal relationship between diet and health

Profile

Prof. Ryoko KATAGIRI

Prof.  Ryoko KATAGIRI

Professor, Faculty of Informatics, Chiba University / Graduate School of Informatics, Chiba University. She graduated from the School of Medicine, Chiba University in 2008. In 2013, she entered the Medical Science Graduate Program, The University of Tokyo. From 2017, she served as a Project Researcher and then as Section Chief at the Division of Epidemiology, Center for Public Health Sciences, National Cancer Center. She has been in her current position since 2024. Ph.D. (Medicine).

Addressing Social Issues Through Data Science

View Another Research